The major objective of this epidemiological project are: 1) to determine the prevalence, types, and natural history of behavior problems arising in the first three years of life, and 2) to examine the causal relationship between stressful life events and other family environmental variables and behavior problems in early development. The sample is taken from several birth cohorts of children born in families in permanent residence on the island of Martha's Vineyard, Massachusetts. Since 1977 the prevalence and natural history of behavior problems in 3-year-old children have been determined and a follow-up of these same children was carried out in 1981. Beginning in 1980, a new phase of the study began by collecting family environmental data and temperament characteristics of the index children in three cohorts of children born between 1978 and 1980. This prospective study will provide a rigorous test of the hypothesis that events in the early family enivronment are causally related to psychiatric disorder in children. The significance of the project relates primarily to establishing norms on variables important in early emotional and behavioral development and in establishing an empirical base of knowledge for the primary prevention of child psychiatric disorder.